Decision & Risk Analysis

Jenn Bonilla

Global Head of Business Insights, Operations, and Established Products
F. Hoffman-La Roche Ltd.

Pharmaceutical Decision-making in the World of Biosimilars

Biologic drugs have revolutionized the treatment of many diseases. Biosimilars are follow-on products that are able to demonstrate a high degree of similarity to an approved biologic. Currently, there are biosimilars approved for immunology and supportive oncology indications but there are as yet no biosimilars to drugs that treat cancer; however, these are expected very soon. It is unknown how physicians, payers, and patients will take up biosimilars in place of originator biologics in this case.

This talk will outline:

-Rationale for using probabilistic instead of scenario models
-Analytic structure to simplify understanding of key uncertainties
-How outputs may inform understanding and decision making

The entrance of biosimilar competition is potentially of great value for society. By supporting all stakeholders’ understanding of value-drivers, we can better ensure the market develops such that greater investment in this area of medicine is encouraged.

Ritske van der Meer

Senior Consultant
Strategic Decisions Group

Co-presenting with Jenn Bonilla


Ritske van der Meer, a senior consultant, has 15 years of experience in life sciences, technology, manufacturing, and oil & gas. He is passionate about taking a decision science perspective on R&D, product portfolio and strategy development for new and existing businesses and assets.
In his 5 years with the firm, Mr. van der Meer has served clients across Europe and US. His clients include companies in BioPharma, Energy and Technology. In these engagements, he has led client teams to develop creative strategies and to take calculated risks. He is experienced in decision framing, in structuring complex problems, creating creative and robust solutions to strategic challenges and developing decision capability building programs. He always keeps teams on track with maintaining a strong decision focus.

Prior to joining SDG, Mr. van der Meer was a senior decision support analyst with ASML in the Netherlands. Mr. van der Meer is Founder and Secretary of the European Decision Professional Network Foundation (EDPN), a growing professional community in Europe, where he co-organizes face-to-face decision science practice events every year and a European Conference every second year.

He holds an MS in econometrics from the Tilburg School of Economics, an MS in innovation sciences from Eindhoven University of Technology, and a BS in aeronautical engineering from Haarlem University, which today is part of Delft University of Technology.

Stefan E. Karisch

Boeing Commercial Aviation Services

 Towards a World of No Surprises – Reducing Uncertainty in Aviation

Operating an airline is an extremely complex undertaking and faces constant uncertainty. In order to successfully dispatch hundreds of aircraft, schedule thousands of air crew, operate up to one hundred thousand flights per month, and transport tens of millions of passengers per year, detailed planning and near-flawless execution are critical.

Unfortunately executing to plan isn’t easy as small to large unforeseen issues occur moment-to-moment every day; from a pilot calling in sick to an airplane with a maintenance issue or a network disruption due to an incoming snow storm. Obtaining a glimpse into the future, gaining more time to make data driven decisions, is a clear advantage to any operation of this day-to-day complexity.

We believe analytics can deliver the advantage – a peak into what may occur. This will provide airlines with more time to adjust and manage to situations rather than react to them. We envision an analytics empowered future; where the aviation industry rarely experiences unplanned or unforeseen events and where it has attained significant improvements in efficiency, economy, performance and safety. I will give examples how analytics tools, methods, and applications are already today realizing this future vision.


As director of Analytics, Stefan Karisch is responsible for the development and application of data analytics capabilities to improve decision making across Boeing Commercial Aviation Services and enhance the experience of Boeing customers globally.

Stefan began his career in 1998 with Carmen Systems, which was purchased in 2006 by Jeppesen, a Boeing subsidiary. He has worked in a variety of commercial and technical leadership roles for Carmen Systems, Jeppesen, and Boeing in Gothenburg (Sweden), Montreal (Canada), Denver, and now Seattle. Under his leadership, Jeppesen excelled in fact-based decision making for which the company was awarded the prestigious INFORMS Prize in 2010 by the Institute for Operations Research and the Management Sciences. He also led the definition and development of new operational efficiency solutions based on advanced analytics for the commercial, military, and business aviation markets.

Prior to his career in industry, Stefan held academic positions in Denmark and Austria. He received a doctorate and a master’s degree in mathematics from Graz University of Technology (Austria) and a master’s degree in mathematics from the University of Waterloo (Canada). In addition, Stefan served as vice president of the Institute for Operations Research and the Management Sciences (INFORMS) and as president of the Airline Group of the International Federation of Operational Research Societies (AGIFORS), and he is the current president of the Analytics Society of INFORMS.

Detlof von Winterfeldt

Tiberti Chair for Ethics and Decision Making, Viterbi School of Engineering
University of Southern California

Identifying, Structuring, and Comparing Objectives of Terrorists

A key question of terrorism risk analysis is: What do terrorists want? To answer this question, we developed a methodology using principles of decision analysis to identify, structure, and compare objectives of terrorists from publically available materials. We applied this methodology to identify and structure the objectives of Al Qaeda, ISIL and Hezbollah and, in the process, found interesting similarities and differences. The results of this work was used in two real-time simulations by the Department of Defense to develop messages to counter ISIL and its followers. Co-Authors: Detlof von Winterfeldt, Epstein Department of Industrial and Systems Engineering, Viterbi School of Engineering, University of Southern California; Johannes Siebert, Operations Management, Faculty of Law, Economics, and Management, University of Bayreuth, Universitätsstr; Richard S. John, Department of Psychology, Dornsife College of Letters, Arts, and Sciences, University of Southern California


Detlof von Winterfeldt is the Tiberti Chair for Ethics and Decision Making at the Viterbi School of Engineering of the University of Southern California (USC) and a Professor of Public Policy of USC’s Sol Price School of Public Policy. In 2003 he co-founded the National Center for Risk and Economic Analysis of Terrorism Events (CREATE), the first university-based Center of Excellence funded by the US Department of Homeland Security. His research interests are in the foundation and practice of decision and risk analysis as applied to the areas of technology development, environmental risks, natural hazards and terrorism. He published widely in these areas. In 2000 he received the Ramsey Medal for distinguished contributions to decision analysis from the Decision Analysis Society of INFORMS and in 2012 he received the Distinguished Achievement Award from the Society for Risk Analysis.

Michael Runge

Research Ecologist
U.S. Geological Survey, Patuxent Wildlife Research Center

Using Multi-criteria Decision Analysis to Explore Management Options in the Grand Canyon

The Colorado River ecosystem between Glen Canyon Dam and Lake Mead is important to many stakeholders, including: American Indian Tribes, whose creation stories begin there; citizens of the seven western Colorado River Basin states, who rely on the river for water and electricity; visitors from around the world, who raft, hike, camp, and fish in the extraordinary wilderness; and environmental advocates, who seek to conserve the species and ecosystems of this unique place. In October 2016, the Bureau of Reclamation and the National Park Service published a Final Environmental Impact Statement (EIS) concerning the long-term management of water releases from Glen Canyon Dam, and in December 2016, the Secretary of the Interior signed a Record of Decision to choose the preferred alternative from the EIS. In developing this EIS, the lead agencies used two formal decision analysis methods, multi-criteria decision analysis and the expected value of information, to evaluate seven alternative strategies against the 18 performance metrics and to evaluate the influence of uncertainty. As part of the multi-criteria decision analysis, stakeholder agencies were invited to participate in a swing-weighting exercise to understand their range of perspectives; the results of the swing-weighting exercise were combined with the evaluation of the alternatives to complete the decision analysis. Because of the large number of performance metrics, novel analytical approaches were needed to understand the variation in weightings across stakeholders; a principal components analysis of the weightings proved valuable. The choice of a preferred alternative was sensitive to the value-based judgment about how to place relative weight on the resource goals. The value of information analysis examined the influence of several hypotheses that described structural uncertainty in the consequence analysis, as well as the influence of several sources of stochastic variation. As an analytical technique, the partial value of perfect information was used to examine the relevance of individual sources of uncertainty to the management decision, and hence, to suggest hypotheses that would be valuable for testing in an experimental design. As it turned out, the rankings of the alternatives were not sensitive to the critical uncertainties that were evaluated, nor to the stochastic uncertainties. This surprising result provides insight into how the influence of uncertainty can be attenuated in a multi-objective setting. Many natural resource management problems have similar features—a large set of stakeholders, many competing objectives, numerous sources of uncertainty, and a wide array of potential management alternatives. Framing these problems is challenging, and analysis may require a wide set of tools. This project demonstrates the value of formal decision analysis methods in the context of an Environmental Impact Statement, to integrate value judgments and scientific evidence in the evaluation of natural resource management alternatives.


Michael C. Runge is a research ecologist with the U.S. Geological Survey, Patuxent Wildlife Research Center, where he was worked since 1999. His research focuses on the use of decision theory and population modeling to inform wildlife management, with particular emphasis on the formal application of adaptive management. Most of his research involves collaboration with Federal management agencies (U.S. Fish and Wildlife Service, Bureau of Reclamation, National Park Service, National Marine Fisheries Service, and others). He has worked on projects with migratory birds, National Wildlife Refuges, endangered species, and marine mammals. He is co-chair of the Polar Bear Recovery Team, work for which he received the US Fish and Wildlife Service’s 2015 Recovery Champion Award. He co-designed the “Introduction to Structured Decision Making” and “Adaptive Management” courses for the National Conservation Training Center, and co-leads the joint USGS/FWS Structured Decision Making Workshops. Mike received a B.A. in biology and philosophy from the Johns Hopkins University, an M.A.T. (Master of Arts in Teaching) in biology from Spalding University, and a Ph.D. in wildlife science from Cornell University. With his wife and daughters, he has climbed all 46 of the high peaks in the Adirondacks.

Faker Zouaoui

Chief Analytics Officer

 Driving Customer Engagement through Personalized Content Recommendation

 Positive user engagement with your mobile app helps deliver customer loyalty and satisfaction. Users have high expectations of the mobile experience and brief attention spans, so careful personalization is crucial for effective communication and engagement with them. Understanding users’ unique needs, preferences, and app interactions is key to personalizing your app experience. In this presentation, I will focus on how to use machine-learning techniques to provide personalized content recommendations and engage with customers the best way. Machine learning leverages vast amounts of structured and unstructured data to generate previously-hidden actionable insights that drive higher customer lifetime value. This presentation will also highlight the technology used to enable these capabilities and provide few examples from the new mobile user experience.


Faker Zouaoui is the Chief Analytics Officer for Asurion, responsible of the enterprise data and analytics strategy, and delivering analytics solutions that drive financial growth and improve customer experience.

Prior to Asurion, Dr. Zouaoui held several leadership positions at Sabre in research, product marketing, and technology where he led the delivery of various travel industry solutions like the ability to shop for the cheapest airline fares used globally by several online travel agencies.

Dr. Zouaoui holds a Ph.D. degree in Operations Research from North Carolina State University. He published papers in the areas of simulation, pricing, revenue management, and travel retailing.